Memory Search Mcp

This is a persistent memory layer for Claude that stores and retrieves conversation context using semantic search. It tracks what the AI cares about through weighted episodes, links related memories into temporal chains, and consolidates information over time into a queryable knowledge base. You'd reach for this when you need Claude to remember context across sessions without re-explaining background, or when building assistants that need to recall past interactions, preferences, or decisions. The system is built with EU AI Act compliance features baked in, including bias detection and transparency tooling. Install via pip or Smithery, and it plugs in through standard MCP stdio transport.